Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 10514
    Prev
    Next
    INDEX
    Explanations

    words related to motivation and encouragement

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "inspire" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rolled
    -0.81
     processing
    -0.69
     Closed
    -0.68
     settlement
    -0.66
    ulo
    -0.66
     deposits
    -0.64
    termination
    -0.64
     transactions
    -0.64
     transaction
    -0.62
    bage
    -0.62
    POSITIVE LOGITS
     inspire
    3.41
     inspires
    2.69
     inspiring
    2.26
     motivate
    1.91
     inspirational
    1.85
     inspired
    1.71
     inspiration
    1.71
    Insp
    1.70
     Insp
    1.60
     motiv
    1.52
    Activations Density 0.023%

    No Known Activations